Skip to content

umutisik/gan-experiments

Repository files navigation

Generative Adversarial Network Experiments

Notebooks experimenting with Generative Adversarial Networks. Using Keras and Tensorflow.

The convolutional model (Convolutional_GAN.py) worked best.

Following:

  • Goodfellow et. al, Generative Adversarial Nets
  • Radford et. al., Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

Random generator outputs:

Non-picked generator outputs


Discriminator Favorites:

Discriminator Favorites


Transitions in the generator input space:

Transitions

License

MIT

About

Experiments with generative adversarial networks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published